> ## Documentation Index > Fetch the complete documentation index at: https://koreai-agent-management-platform-dev.mintlify.app/llms.txt > Use this file to discover all available pages before exploring further. # Standard Responses [Back to NLP Topics](/ai-for-service/automation/natural-language/nlp-topics) Standard responses are pre-defined NLP interpreter messages the AI Agent displays based on events, triggers, user inputs, or system conditions. You can customize all defaults and add channel-specific overrides. Go to **Automation AI > Conversation Intelligence > Conversation Management > Standard Responses**. For each response category, the trigger or user input is listed alongside the default response(s). The NLP interpreter randomly selects from multiple responses at runtime. *** ## Response Categories | Category | Description | | --------------------- | ---------------------------------------------------------------------------------------------------- | | **Statements** | Responses triggered by AI Agent or user actions — for example, task cancellation or invalid input. | | **Queries** | Questions the AI Agent asks for more information — for example, *Which field do you want to change?* | | **Errors & Warnings** | Messages displayed during errors — for example, *End date cannot be before the start date.* | | **Questions** | General questions posed to the user that don't fit other categories. | | **Choices** | Yes/No prompts — for example, *Would you like to receive push notifications?* | | **Greeting** | Initial responses when the app starts or when a user returns after inactivity. | *** ## Statements Responses triggered by actions from the AI Agent or user. | Message | When It Appears | | --------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- | | When the user confirms to close the conversation in a language and switch to another language | User ends a conversation in one language and starts the next in a different language. | | When only field prompt is present in the OOB | Internal field prompt for action and alert tasks. Not recommended for customization by enterprises. | | When language preference is set | User switches language using *cheat lang* or picks from language options. | | When task is not accessible to user | Developer tries to access an app task they don't have permission for. | | User confirms alert setup | An alert is set up successfully (Alert Task only). | | User discards/abandons the current activity | User says *discard*, or the task fails due to max retries, invalid LoV, or cURL error. | | User implies to end the conversation | User says *bye*, *good night*, etc. Shown when the Follow-up Intents list is presented at dialog end. | | Location provided by user is invalid | AI Agent cannot validate input for a Location-type entity. | | Amending an Alert field | User says *Yes* to amend alert parameters (Alert Task only). | | Amending a task which has no fields | Task has no entities to amend (Alert, Action, and Information Tasks). | | Ask at least one optional field | User did not provide any optional field value when at least one is required (Action Task only). | | Topic retry limit exceeded | Developer mode only. Dialog looped through the same nodes beyond the allowed limit. | | Universal fallback response | AI Agent cannot generate a response for any reason, or maximum node execution limit exceeded. | | Server returned authorization failure | Authorization failure occurred during script or API execution. | | User selects a valid option from a list | User selects an action/alert/info task from the listed options. | | User selects an action | User selects an action task; AI Agent confirms and asks for required parameters. | | Alert is identified, ask for alert time | Displayed when Trigger Interval is set to *At a specific time*. | | Alert is identified, ask for alert time interval | Displayed when Trigger Interval is set to *Every X minutes*. | | User selects an alert | User selects an alert task; AI Agent confirms and requests required parameters. | | Task discarded | User types *discard* mid-task. | | App choosing the default account | Shown during alert amend when AI Agent cannot parse the user's response. | | Confirmation Dialog for Task consolidated | AI Agent confirms captured parameters before executing an Action/Info task. | | Confirmation Dialog for Alert consolidated | AI Agent confirms all details before setting up an alert (if confirmation is enabled). | | Confirmation Dialog for Alert consolidated without fields | Alert has no fields; AI Agent confirms before execution (if confirmation is enabled). | | The alerts will be muted | Developer has muted alert notifications. | | Confirmation Dialog when alerts are muted | Shown when *Mute alert notifications* is enabled for an RSS alert. | | Alert frequency description | Confirmation of alert query frequency for web-service alert tasks. | | Not understanding the input at confirmation | User's response during alert frequency confirmation does not match the expected input. | | No results found for the search string | Typeahead field returned no results. | | Successful task setup | Alert task set up successfully. | | I am not sure what you mean | One or more low-confidence intents found. Usually shown as FAQ suggestions. | | Ambiguity in intent recognition | AI Agent failed to identify user intent; prompts user to rephrase. | | User has selected daily alert — ask for frequency or specific time | Both *frequency interval* and *specific time* options are available. | | App needs to present filter options | Multiple filters available during alert setup. | | App needs to present filter options with exit option | Same as above, but includes an option to skip. | | User finished configuring filters | Filter configuration for an alert task completed. | | User asks *Who am I* | User types *Who am I*. | | User asks for help | User types *Help*; AI Agent lists available tasks. | | Skipped values | Part of confirmation message for Action/Info tasks — shows skipped (optional) parameters. | | Ask user to change dependent field | A dropdown depends on another field, but the Platform cannot fetch data from that field. | | When there is no data | User chooses to change account but responds *no* to the AI Agent's confirmation. | | When date given in invalid format | User enters an unrecognizable date format. Valid formats: YYYY-MM-DD, DD-MM-YYYY, 1 Jan 1982, *Next Friday*. | | Follow-up sentence post task discard | AI Agent confirms task was discarded and offers to restart. | | Confirmation Dialog for task with skipped fields | Confirmation before executing an Action/Info task when some optional fields were not provided. | | Listing all tasks | Displayed when user types *Help* and no task matches. | | Listing all alerts | Displayed when user types *Help* and the app has only alert tasks. | | Listing all FAQs | Displayed when user types *Help* — includes Knowledge Graph collections. | | Listing all alerts after tasks | Shown when app only has alert tasks and intent is not identified. | | Listing all alerts after tasks and FAQs | Shown when all task types are configured and user types *Help*. | | Lists all actions and FAQs | No matching task found; lists actions and FAQs. | | Lists all alerts and FAQs | No matching task found; lists alerts and FAQs. | | When app has no tasks | Developer tries to run an app without any tasks. | | When a server call fails | Webhook or service node call failed; task is discarded. | | User prompt when a task is discarded due to server call failure | API call did not complete successfully. | | Specific app is not available | App is not yet published. | | Field title and value in confirmation | Part of the confirmation step — shows captured field values for alert, action, or info tasks. | | Recursive dialog found | AI Agent recursively called the same dialog and exceeded the limit (100). | | Loop limit reached for a node | Same node invoked repeatedly beyond the allowed number of times. | | Message component doesn't have a message | Message or Service node returned empty content. | | User says thanks | User thanks the AI Agent. | | User says OK, fine, yes | Shown after task completion when the user acknowledges positively. | | User says no, nope | Small talk response to negative acknowledgments. | | User says great, awesome | Shown after task completion when the user responds positively. | | User says that helped, that was useful | Shown after positive feedback (great, cool, fair enough, that helped). | | User says I am done | Shown when user indicates they are finished (done, that's all, that's it). | | Successfully added filter | A filter was added to an alert. | | After successfully adding a filter, showing filter options again | Shown when a filter is applied to an alert (Alert Task only). | | User gave invalid input for filter | Filter name doesn't match any configured filter. | | Resume paused task | User chooses to resume on-hold task. Shown only when Resume option is *Notify the user with a message that the on-hold task is being resumed*. | | Notify the user that we have canceled the paused task | User chose not to resume the on-hold task during confirmation. | | Notify the user that we have canceled the current task to switch to another | Hold option is *Discard the current task and switch to new*. | | When an interruption is identified but hold is not allowed | New intent detected during task execution; new intent added to follow-up list (current task continues). | | Service call timed out | Third-party API call (webhook or service node) timed out. | | Dialog has no transitions | Dialog task has only an Intent node and no further flow defined. | | Manage account | User says *manage accounts*, *add phone number*, *delete account*, etc. | *** ## Queries Questions the AI Agent asks users for more information. | Message | When It Appears | | ---------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------- | | Ask if user would like to switch to a new language (multiple detected) | Multiple new languages detected mid-conversation; user asked to select preferred language. | | Ask if user would like to switch to a new language (single detected) | A different language detected; user offered option to switch or continue. | | Follow-up sentence post task execution or discard | Task completed or discarded; user offered option to restart. | | Confirmation before discard | User types *Discard* or *Discard All* during task execution (applies to select clients). | | User types *Change* or similar words | User changes their mind during alert filter setup; alert setup is discarded. | | User prompted to enter text to search | Typeahead field prompt (Action, Alert, Info tasks). | | Ask the user to enter alert frequency | Frequency can be interval or specific time (Alert only). | | Ask the user to enter alert interval | Frequency is an interval, e.g., every 4 hours (Alert only). | | Ask the user to enter alert time | Frequency is a specific time, e.g., daily at 9:00 am. | | Alert identified — ask for frequency | Prompted based on schedule options defined by the alert developer. | | Prompting user to enter name for alert instance | User installs alert in Kore messaging rooms. | | User needs to select from a static dropdown | Filter field type is *static dropdown*. | | User needs to select from a search results list | Filter field type is *dynamic dropdown*. | | Request for email address for webhook instructions | Webhook alert setup — developer email needed for configuration instructions. | | Query to get the alert instance description | User prompted to provide a description during alert setup. | | Tell me the alphabet or unique word for the action | Displayed before a list of options when no custom template is used. | | I am not sure what you mean after an invalid choice | User's follow-up doesn't match any of the initially identified ambiguous alerts. | | Asking user to pick an alert | A list of alerts is shown; user prompted to select one. | | Asking user to pick an alert after an error | User selected an invalid option from the alert list. | | When user responded with invalid answer to confirmation dialog | User said they want to amend but the AI Agent didn't understand what — confirmation repeated. | | Typeahead field prompt (first field in task) | First field in task is a typeahead type. | | Ask user what to rename the alert | User chose to change/amend the alert name. | | After user amends a parameter — ask if anything else to change | User amended a parameter; AI Agent asks if further changes are needed. | | No results found for search string — ask to try again | Typeahead field query returned no results. | | Follow-up sentence post discard — Part 2 | At alert confirmation, user said *no* twice. | | Listing ambiguous tasks | User prompted to pick a task from a list; response led to ambiguity. | | Ask if user wants to continue the paused task | Shown after the interrupting task completes (Resume option: *Get confirmation before resuming*). | | Ask if user would like to switch | Hold & Resume is set to *Allow the end user to select the behavior*; AI Agent confirms before switching. | *** ## Errors & Warnings Messages displayed when errors or validation issues occur. | Message | When It Appears | | ---------------------------------------------------------- | ------------------------------------------------------------------------------------------------------ | | Agent transfer fallback response | Agent unavailable, wait time exceeded, or agent transfer failed. | | User exceeds maximum retries for an entity (task on hold) | User exceeded max retries while another task is on hold. | | Dialog task associated to FAQ is not available | The dialog linked to a FAQ was suspended, deleted, or not assigned to the user. | | User needs to authorize or re-authorize | End-user authorization or re-authorization is required. The authorization URL is appended dynamically. | | User authorization has failed or is yet to authorize | Task requires authorization; AI Agent provides an authorization link. | | User's intent recognized, but authorization fails | Service call needs authorization and no account has been authorized yet. | | User authorization failed more than 3 times | User entered wrong credentials three times in a row. | | User provided value is not unique | User's input causes ambiguity when selecting from a partially matched intent list. | | App did not find any options based on user's choice | Option set is empty, or no account credentials are available. | | Discarding task — no options found for list or lookup | User input not found in service call response (dynamic dropdown or typeahead field). | | Task identified, but server call failed | Kore account for the third-party API expired or failed. | | User enters invalid email | Invalid email address entered for an alert, action, or info task. | | User enters invalid country | Invalid country name entered. | | User enters invalid URL | Invalid URL entered. | | User enters invalid phone number | Invalid phone number entered. | | User enters invalid number | Non-numeric input entered for a number field. | | Invalid date format | Invalid date format entered. | | Missing date | No date provided for a date field. | | Ask user to enter time for a field | Time missing for a DateTime entity in a dialog task. | | Ask user to enter time | User provided only a date for a DateTime entity (e.g., scheduled appointment). | | Invalid time — ask to try again for a field | User's time input for a DateTime entity is missing or invalid. | | Invalid time — ask to try again | User entered an invalid time value (e.g., Aug 29 at 25). | | Missing date/time | Error prompt for DateTime entities in Actions and Alerts. | | You have chosen a wrong choice | Sub-response used when a user selects an invalid option from a list. | | Cannot understand your request | AI Agent fails to identify intent; lists available tasks and FAQs. | | App is selected, user asked for reauthorization | API call failed due to expired credentials. | | Listing all tasks after wrong choice | User selected an invalid option; app has only Actions/Dialogs/Info tasks. | | Listing all alerts after wrong choice | User selected an invalid option; app has only Alert tasks. | | Listing all FAQs after wrong choice | User selected an invalid option; app has only FAQs. | | Listing all FAQs — not understanding request | App has only a Knowledge Graph and fails to understand intent. | | Listing all actions — not understanding request | App has only Action tasks and fails to understand intent. | | Listing all actions and alerts — not understanding request | App has only Action and Alert tasks and fails to understand intent. | | Listing all actions and FAQs — not understanding request | App has only Actions and a Knowledge Graph and fails to understand intent. | | Listing all alerts — not understanding request | App has only Alert tasks and fails to understand intent. | | Listing all alerts and FAQs — not understanding request | App has only Alert tasks and a Knowledge Graph and fails to understand intent. | | Listing all tasks — not understanding request | App fails to understand intent; shows all available tasks. | | Listing all tasks and alerts after wrong choice | User selected an invalid option; app has tasks and alerts. | | Listing all alerts, tasks, and FAQs after wrong choice | User selected an invalid option; app has all task types. | | Lists all actions and FAQs after wrong choice | App has only Actions and a Knowledge Graph; user selected an invalid option. | | Lists all alerts and FAQs after wrong choice | App has only Alerts and a Knowledge Graph; user selected an invalid option. | | Task discarded — user exceeded maximum allowed attempts | User entered an invalid entity format five times. | | Entity component in dialog has no error prompt | Entity node has no error prompt defined. | | Entity component in dialog has no message attribute | Entity node has no message prompt defined. | | User input exceeds maximum sentence or word limit | Input exceeds the allowed character or word count. | *** ## Questions General questions the AI Agent poses to users that don't fit other categories. | Message | When It Appears | | ---------------------------------------------- | ------------------------------------------------------------- | | When user input matches and frequency is asked | User input led to ambiguity while setting an alert frequency. | | Ask user to input alert frequency | Prompt for frequency at which the alert task will run. | *** ## Choices Yes/No prompts used during alert and task setup. | Message | When It Appears | | ----------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- | | Search results displayed (type-ahead search) | User input triggered a typeahead search in an alert filter (Alert Task only). | | Confirming alert frequency | App confirms current alert trigger frequency with the user. | | Repeat confirmation of default alert frequency | User input doesn't match the expected frequency response. | | Confirmation to set up alert with frequency | Confirm the trigger frequency for an alert. | | Prompting user to set up filters | App asks whether the user wants to configure alert filters. | | Prompting user to set up filters as soon as alert is identified | Same as above — triggered immediately on alert identification. | | Asking user to set up push notifications | User wants to amend the alert notification setting. | | Repeat — asking user to set up push notifications | User's response to notification setting doesn't match yes/no. | | Amending the description | User wants to change the alert description. | | Amending the frequency | User wants to change the alert trigger frequency. | | Amending notification settings | User wants to change whether notifications are enabled. | | Prompting user to choose days of the week | User selecting Weekday/Weekend for trigger frequency. | | Query to choose interval for frequency | User prompted to select a frequency interval. | | Query to send webhook setup instructions via email | App checks if developer wants webhook configuration instructions emailed. | | Do you want to change the alert name? | User wants to amend the alert name. | | Query to choose operator while setting up a filter | User prompted to select an operator for an alert filter. | | Options to pick frequency for alert notification days | All interval options available (Daily, Weekday, Weekend). | | When the user needs to choose from provided choices | User prompted to pick from a dropdown in alerts or actions. | | Confirm if user wishes to reuse a pre-saved context | Prompted when a pre-saved context is available for reuse. | | Show possible intervals for alert schedule | User entered a specific time, but only intervals are valid. | | Show closest frequency for alert schedule | User's input doesn't match available frequencies; nearest option shown. | | Show closest time for alert schedule | User's input doesn't match available times; nearest option shown. | | When changing a field | User selected *No* during alert parameter confirmation. | | Prompting to set up filters after alert identified | User provides input to set up an alert that has filters configured. | | Ask if setup instructions should be sent via email (first prompt) | First prompt in alert setup is to ask about emailing setup instructions. | | While choosing options | Shows options from the Platform for a typeahead entity in an action or alert. | | After selecting task — ask about schedule | Alert selected from list; AI Agent asks how often the user wants to be notified. | | Wrong choice for task schedule | User entered invalid frequency for an alert; AI Agent asks again. | | After selecting task — ask for field value | First field in an action or alert is a dropdown. | | User says *no* for alert setup confirmation | User chose not to proceed with the alert configuration; offered to amend. | | After selecting task — user selects wrong choice in dialog | User entered an invalid choice from a non-LoV list. | | User enters wrong choice for alert frequency | Alert frequency value doesn't match options defined by the developer. | | User enters wrong value for a dropdown | Input doesn't match any value in the List of Values. | | List ambiguous LoV values | Multiple partial matches for a List of Values entity; user asked to select. | | User enters invalid choice for an action | User's response doesn't match the presented task list. | | Ambiguous date | User entered an ambiguous date (e.g., 6/9); both interpretations shown. | | Ambiguous airport | User entered an airport name that matches multiple airports (e.g., London). | | Ambiguous location | Geocode service returned multiple location possibilities. | | More than one account to pick from | User has multiple third-party accounts; asked to select one. | | Intent recognized, but more than one account to pick from | Same as above, when intent is identified. | | Custom lookup returned more than one value | Multiple results returned from a custom list query. | | Chosen list is empty after selecting a valid choice | Dependent dropdown has no options because the parent field value returned empty results. | | User selects an invalid choice (dependent dropdown) | Same as above; user's value for the dependent field is invalid. | | Present follow-up tasks at end of dialog | Shown at end of dialog when Follow-up Task setting is *Yes, ask the user to select a follow-up task*. | | Ask user to select entity to amend (ambiguous entity value) | Multiple entities share the same value during dialog execution; user asked to clarify which to amend. | | Max on-hold quantity exceeded — ask user to cancel a task | User tried to put more tasks on hold than the configured maximum. | | User provides intent when prompted for String/Description entity | User input matched an intent when prompted for a string/description entity; AI Agent asks how to proceed. | *** ## Add and Edit Responses To add a new standard response: 1. Select a **category tab** and hover over the target situation. 2. Click **Add Response**. 3. In the **Add Channel Overrides** dialog: * Select a **channel** (or **Any Channel** for all channels). * For channel-specific responses, select a **task** and optionally a **field**. add channel override 4. In the **Response** editor: * **Simple Mode** — Enter plain text with optional formatting (bold, italics, hyperlinks, lists). * **Advanced Mode** — Enter JavaScript for custom channel-specific messages. Use `response.message` for the default message string and `response.choices` for the options array. 5. Click **Save**. ### Channel Override Examples **Slack:** ```js theme={null} var message = {}; var attachments = []; var attachment = { 'text': "Exception occurred during API execution", 'color': '#FFCC00', 'attachment_type': 'default' }; attachments.push(attachment); message.attachments = attachments; print(JSON.stringify(message)); ``` **Facebook:** ```js theme={null} var message = { text: response.message }; var buttons = []; for (var i = 0; i < response.choices.length; i++) { var choiceItem = response.choices[i]; if (choiceItem.length > 20) { choiceItem = choiceItem.substring(0, 18) + '..'; } buttons.push({ content_type: 'text', title: choiceItem, payload: choiceItem }); } message.quick_replies = buttons; print(JSON.stringify(message)); ``` **SMS:** ```js theme={null} var message = {}; var indexArray = ['a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z']; var text = response.message + '\n'; for (var i = 0; i < response.choices.length; i++) { text = text.concat(indexArray[i] + ') ' + response.choices[i] + '\n'); } message.text = text; print(JSON.stringify(message)); ``` **Email:** ```js theme={null} var message = {}; var text = response.message + ' '; for (var i = 0; i < response.choices.length; i++) { text = text.concat('' + response.choices[i] + '
'); } message.text = text; print(JSON.stringify(message)); ``` ### Customizing Ambiguity Display with `getChoices()` Use `getChoices()` to retrieve ambiguous intent choices and display them in a custom format (for example, as buttons): ```js theme={null} var info = getChoices(); var message = { "type": "template", "payload": { "template_type": "button", "text": "Did you mean", "subText": "You can now customize response for did you mean", "buttons": [] } }; for (var i = 0; i < info.length; i++) { message.payload.buttons.push({ "type": "postback", "title": info[i], "payload": "payload1" }); } print(JSON.stringify(message)); ``` Built with [Mintlify](https://mintlify.com).